← Back to community benchmarks

Qwen3.6-35B-A3B-oQ8

M1 Ultra (64c) · 128 GB · 8bit · 2026-04-16
Performance
8k
tokens
1,092
PP tok/s
19.4
TG tok/s
7499
TTFT (ms)
45.0
Peak mem (GB)
Hardware
Chip M1 Ultra (64c)
Memory 128 GB
GPU Cores 64
Software
oMLX v0.3.5
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 798.1 54.3 35.4 GB view
4k 1,064 10.7 43.2 GB view
8k 1,092 19.4 45.0 GB current
Batching Results
Batch Size TG tok/s Speedup
54.3 1.00×
32.1 0.59×
71.9 1.32×